Linguistics ‘Hall Of Shame’ 1

Hi again, everybody!

At Bob’s suggestion, I’m presenting a new series of blogs on specific ‘fringe’ writers (idiosyncratic individuals rather than members of ‘movements’) on language matters (in his words a ‘Hall Of Shame’, though some of these writers are more worthy of attention than this terminology might suggest!).


One example of a deep-thinking person who developed his own idiosyncratic linguistic theories (and issued attacks on the mainstream of linguistics) is the psychologist and amateur logician John Trotter. Trotter was a psychology lecturer in Australia, and in the 1970s-80s he developed radical views on the logical and structural nature of language, and incorporated these into papers offered for publication and into his teaching. His most central focus was on the philosophical aspects of these matters, and one of his main foci within linguistics involved the development of links between the formalism of Chomskyan grammars on the one hand and the ‘predicate calculus’ of logic on the other. If Trotter’s main ideas are valid, much of the basis of linguistic theory and indeed some important aspects of contemporary thought on logic must be mistaken.

Trotter’s papers were rejected by editors and reviewers whom he regarded as inadequately informed, and he was allegedly discouraged from presenting his views to students. After that time he operated as a private scholar. He & I interacted at some length in the late 1990s, but his dogmatic approach to many of his key points does not encourage scholarly attention. (In addition, some of his claims regarding e.g. possible improvements in automatic translation appear to be contradicted by familiar linguistic facts.)

Some of Trotter’s main points involve direct criticisms of mainstream concepts. For instance, he holds that some key mainstream linguistic concepts such as ‘allophone’ and ‘phoneme’ should NOT be used; new formulations of the matters in question should be adopted. This stance involves his rejection in its mainstream form of the ‘emic/etic’ contrast (as in phonemic versus phonetic), which he instead treats as essentially a type-token relationship; for example, each phoneme is seen as a type and each of its allophones as a token of this type. Trotter’s objection at this point is associated in turn with his opposition on philosophical (ontological) grounds to some types of linguistic expression used to express the relationship between types and tokens (for example, he objects to definite descriptions such as the dodo as used to refer to a type, as in the dodo is now extinct). However, whatever might be the merits of Trotter’s philosophical points, his reasons for rejecting the mainstream formulations of the linguistic concepts in question here (‘allophone’ etc.) appear to involve a) a degree of misreading of mainstream linguistics (an allophone of a phoneme is NOT in fact the same thing as a token of that phoneme; it is itself a type, more specific and at a less abstract linguistic ‘level’, and has its own tokens, namely the individual instances where it occurs) and b) an exaggerated ontologically-based preference for one linguistic formulation of such matters over another (not uncommon in philosophical work based on linguistic facts; see below).

Trotter also rejects (mainly covertly, and in some specific respects) the linguistic non-prescriptivist approach to language adopted by mainstream linguists. He argues that certain kinds of formulaic expression of philosophical interest (for instance the logician’s For all X, X is Y = ‘all Xs are Y’, as in ‘all men are mortal’) are to be deemed ungrammatical even though they are the normal forms used by the relevant native speakers (logicians) in such cases. For all X, X is Y is seen as ungrammatical because there is no determiner such as that or the before the second token of X, which would be required in more everyday styles of English if the sentence were to occur (suitably modified) and to be deemed grammatical. For instance, in any other context one would not say For all lions, lion has paws, or even For every/any lion, lion has paws; the noun lion in the second clause would always have a determiner, as in that lion or the lion. However, this does not imply that it must have a determiner in the style used in the technical philosophical/logical domain in which such sentences normally occur. These sentences are NOT ungrammatical in context in any normal sense of this term (prescriptivist or descriptivist).

Trotter goes on to argue that because these expressions are ungrammatical they are also logically invalid – and that, because the issue at hand is central in discussions of logic, the whole basis of logic is thereby impugned. Indeed, the philosophical under-pinnings of contemporary logic and linguistics are grossly inadequate. Both mainstream linguists and philosophers would deny this. First of all, the linguistic features in question are found in only some languages. Not all languages even require determiners modifying nouns; for instance, Chinese and Russian do not. In fact, there are serious problems (albeit not always adequately acknowledged by philosophers) associated with heavy reliance upon linguistic data in a philosophical context, because the details of the constructions involved vary so much from language to language. And, even if it were accepted that ENGLISH sentences such as those cited were ungrammatical, it would thus be difficult to argue that some formulations were logically invalid on the ground that there was an issue with the grammar of the versions of these formulations as expressed in English but not in all languages. English and similar languages have no special status in this respect.

Indeed, the grammatical and logical statuses of expressions in any given language are very largely independent of each other. Grammatical issues can normally bring the logical status of expressions into question only if they involve the meanings of these expressions, for example by rendering them ambiguous or self-contradictory. Most grammatical anomalies (whatever their origin or the status ascribed to them) generate no significant ambiguity and certainly have no logical consequences. A form such as Jo go home (grammatically non-standard) is normally semantically transparent, and its ‘ungrammaticality’ (in contrast with Jo goes home, etc.) has no logical consequences.

For reading, see John Trotter, System of Rational Discourse: Vols. 1 (A Calculus of Attributes), 2 (Aggregates, ‘Numbers’ and ‘Sets’), 3 (Applied Arithmetic and Probability), 4 (A Calculus of Predictions and Propositions), (Aranda (ACT), 1995-6, and other works; see
bibliography with links to online versions at http://isbndb. com/d/person/trotter_john. html (accessed 2 February 2011).



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: